BuildRAG is a local, production-ready Retrieval-Augmented Generation (RAG) workspace built with Streamlit, Ollama, and Chroma.
It lets you upload documents, index them locally, and chat with an LLM that answers questions grounded in your own data, with sources.
- 📂 Upload & manage PDF / TXT / MD documents
- 🧠 Local vector database with Chroma
- 🔍 Multiple retrieval modes (similarity, MMR, score threshold)
- 💬 Chat interface with source citations
- 🛠️ Debug mode to inspect retrieved chunks
- 🔒 Fully local (Ollama-powered LLM & embeddings)
- UI: Streamlit
- LLM: Ollama (e.g. llama3.1)
- Embeddings: nomic-embed-text
- Vector DB: Chroma
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
ollama serve
streamlit run app.pyollama serve
streamlit run app.py